Using bulky reagents such as 9-BBN, hydroboration is often highly diastereoselective, giving rise to the 1,2-anti diastereomer:
This is easily explained using a steric argument, with the major product resulting from minimisation of allylic strain in the transition state. This is essentially the Houk model, though in the context of hydroboration is often described as the reactive-conformer model.
While a steric argument works in many cases, it isn't fool-proof – increasing the size of a group does not necessarily increase the selectivity as would broadly be expected by the above model.
To rationalise this, Houk has proposed a stereoelectronic component in the transition state – this is perfectly plausible as electronics evidently contribute to the conformational preference of the molecule. What Houk does not expand on is the exact nature of the stereoelectronic effects (this is not unusual, stereoelectronics are often invoked to rationalise unexpected experimental results without justification).
What is the nature of the stereoelectronic component when determining diastereoselectivity of hydroboration and in what situations should it be invoked?